The novel compounds of this invention are N,N-disubstituted .alpha.-(3,5-dialkyl-4-hydroxyphenyl)-.alpha.,.alpha.-disubstituted acetamides ("3,5-DHPA" for brevity) in which one of the substituents on the N atom is a polysubstituted piperidyl group, more correctly referred to as "N-substituted, N-(polysubstituted-4-piperidyl)-.alpha.-(3,5-dialkyl-4-hydroxyphenyl)-.alp ha.,.alpha.-substituted acetamides" hereinafter, referred to as "3,5-DHPIPA" for brevity. These compounds have never heretofore been made because the tertiary alpha-carbon ("alphaC") atom (alpha relative to the phenyl ring) of any reactant from which such a compound might have been derived, is so hindered that it does not permit reaction with an amine to form the amide which, conventionally, one might expect to be formed. By "tertiary" C atom we refer to a C atom bonded only to C atoms. The distinctive feature of our acetamides is that the tertiary C atom, referred to as "the alpha C atom" because it is alpha to the hydroxyphenyl ring, and also alpha to the carbonyl C atom, is disubstituted. In other words, there is a single disubstituted C atom connecting the hydroxyphenyl ring to the carbonyl C atom of the acetamide.
3,5-dialkyl-4-hydroxyphenyl organic compounds, referred to as "hindered phenols" because of the substituents on the ring C atoms flanking the ring C atom carrying the hydroxyl (OH) group, have been of great interest for some time because of their stabilization activity. This interest derived from the discovery that such compounds were excellent antioxidants, this property in turn, being related to the stability of the aroxyl radical represented by the typical structure ##STR1## wherein the "cross" substituents represent t-butyl, written out in greater detail on the 1-C atom. It was, until the discovery of the 3,5-DHPA radical, one of the most stable aroxyl radicals known. This prior art aroxyl radical was referred to as the "blue aroxyl" in a lecture titled "The Blue Aroxyl, The First Stable Oxygen Radical, Its Discovery and Its Properties" given by Eugen Muller, in Lisbon on the 28th of May 1973 and printed in Rev. Port. Quim. 14, 129 (1972). The radical was referred to as "blue" because of the distinctive dark blue crystals obtained by shaking the benzene solution of 2,4,6-tri-tert.butylphenol with potassium ferricyanide in aqueous alkali.
Since the effectiveness of hindered phenols to stabilize an organic material, subject to degradation due to heat, oxygen and light, appeared to be correlated to the stability of the aroxyl radical generated by exposure of the organic material, it seemed likely that modifications in the structure of such hindered phenols, particularly those modifications relating to the substituents on the 1-C atom of the phenyl ring, might provide aroxyl radicals which were more stable than those of the prior art. The quest appeared to devolve upon finding which particular substituent on the 1-C atom provided better stability of the radical than another substituent.
This general approach seemed to have been taken by prior workers in the field, for example by Meier et al. in U.S. Pat. No. 3,247,240, though at the time, it can be assumed they were unaware of the existence of the aroxyl radical. Because he reacted 2,6-di-tert-butyl phenol with an alpha,beta-monoolefinically unsaturated compound such as methyl acrylate, he could never have more than a mono-substituted alpha-C atom, that is, the alpha-C atom could never be a tertiary (that is, fully substituted) C atom. And, of course he could not have provided a substituted acetamide.
U.S. Pat. No. 3,338,833 to Spivack et al., issued soon thereafter, pursued the lead of Meier et al., but with substituted dialkyl-4-hydroxyphenyl amides having an alkylene (`spacer`) group spacing the amide C atom from the phenyl ring. Again, the alpha-C atom could never have more than a single substituent on the alpha-C atom, that is, the alpha-C atom could never be a tertiary C atom. In the reaction schemes he suggests, he specifies the reactant hydroxyphenyl ester or acid chloride as having a (CH.sub.2).sub.n spacer where n is a small whole number, e.g. 1 or 2. Assuming one decided to impute and extend the enablement embodied in the Spivack '833 teachings to a spacer having a tertiary C atom, and, chose to make a hydroxyphenyl amide spaced from the pheny ring only by the tertiary C atom, one would need to have access to the precursor hydroxyphenyl acid having the structure ##STR2##
or its acyl derivative, wherein R.sup.1 and R.sup.2 represent alkyl, cycloalkyl, phenyl, alkyl-phenyl, naphthyl and alakyl-naphthyl which serve to hinder the OH group, and R.sup.3 and R.sup.4 represent alkyl substituents, which acid or acyl derivative could then be used to react with an appropriate amine. This 3,5-dialkyl-4-hydroxyphenyl substituted acetic acid, to be used as a precursor is obtained as disclosed in my U.S. Pat. No. 4,523,032 and is not a prior art compound. The corresponding ester, however, is a prior art compound and may be obtained as disclosed in U.S. Pat. No. 3,455,994 to Knell. I therefore reacted the ester, namely methyl .alpha.-(3,5-di-tert.butyl-4-hydroxyphenyl)-.alpha.-methyl propionate, which is a prior art compound, with tert-octylamine at 150.degree.-160.degree. C. for 4 hr but failed to find any trace of the expected amide with the tertiary alpha C atom. I then continued the reaction at 160.degree.-170.degree. C. for an additional 2 hr and still failed to obtain the expected product.
To make certain this lack of reactivity was not due at least in part to the steric hindrance of the amine group of the tert-octylamine, I chose to minimize any such effect by substituting a long straight chain alkylamine, namely octadecylamine, for the t-octylamine. I then conducted a reaction with octadecylamine and methyl .alpha.-(3,5-di-tert.butyl-4-hydroxyphenyl)-.alpha.-methyl butyrate in an analogous manner and under the same conditions of reaction as those described immediately hereinabove. I found the octadecylamine failed to react with the methyl .alpha.-(3,5-di-tert.butyl-4-hydroxyphenyl)-.alpha.-methyl butyrate. Only a trace of high mol wt material was detected by mass spectrography. This trace of material was identified as the dimer of the amine having a mol wt of 520. The remaining material consisted of the unrected reactants. No amide with a tertiary alpha-C atom was obtained.
Later, in U.S. Pat. No. 3,787,355, Linhart et al., like Spivack '833, chose an alkylene spacer, but changed the amine believing this might provide more effective stabilization. Like Linhart et al., Spivack also pursued the `bulking of the molecule` as a path to more effective stabilization in U.S. Pat. No. 4,049,713, but retained the alkylene spacer between the ester group, or amide group, and the phenyl ring in his hindered hydroxyphenyl alkanoates and amides. Since he had started with a compound which had either only one substituent on the alpha-C atom, or none, the esters he made were prepared via usual esterification procedures from a suitable alcohol and a carboxylic acid derivative of the substituted hindered phenol of interest, so that it is clear he could never have made an ester with a tertiary alpha-C atom.
In U.S. Pat. No. 4,191,683, Brunetti et al. decided upon a polysubstituted piperidyl alkylamine, or a dimer or trimer of it, without linking their compounds to a hydroxyphenyl moiety, clearly indicating that at that time, there was no suggestion that an aroxyl radical might advantageously be combined with a polysubstituted piperidyl alkylamine, irrespective of what linkage might be used. At least with respect to their dimer and trimer, the emphasis was on bulking the molecule.
It is evident that the opportunity to investigate the stability of an aroxyl radical having a disubstituted alpha C atom did not present itself because the alpha C atom could not be disubstituted. This inability is borne out by the efforts of Rosenberger et al in U.S. Pat. No. 4,197,236, who were able to produce an alkyl (methyl) substituent on a tertiary C atom (see example 3) but were forced to provide an alkylene spacer between the tertiary C atom and the carbonyl C atom, to obviate the steric hindrance and allow the reaction to proceed. When they coupled the alpha C atom to the carbonyl C atom of the carboxyl group (see example 1), the alpha C atom was not disubstituted. However, their pursuit of a "bulked-up" acetamide molecule was successful. They were able to provide two hydroxyphenyl groups on the tertiary C atom with the expectation of providing a more stable diradical. This was a different approach towards the same goal we pursued, namely finding and synthesizing a more effective stabilizer. They opted to cope with the destabilizing effect of the tendency of each of the two radicals connected to the single C atom, to form a quinone methide, but benefitted from the bulked up molecule. We chose to use a single hydroxyphenyl radical in which the stability was enhanced by the lone disubstituted alpha C atom connecting the hydroxyphenyl ring to the carbonyl C atom. Because the generation of an aroxyl radical with enhanced stability is the nexus of the activity of our compounds it is evident that they are distinct and different from the dihydroxyphenyl piperidyl compounds of the '236 patent.
Not long afterwards, in U.S. Pat. No. 4,246,198, Rosenberger et al pursued the notion of bulking the molecule even further by forming a dimer or trimer after bulking the substituent on the 1-C atom of the phenyl ring. In each embodiment he provided an alkylene spacer (C.sub.x H.sub.2x) in which x is defined as being 0, 1, 2 or 3, optionally in combination with another such alkylene spacer (C.sub.y H.sub.2y). But the disclosure of the value "0" for x was clearly accidental since there is no suggestion provided to enable one to make such a compound. Nor is there any suggestion that such a compound, which they were earlier unable to make, might now have been made by them in this '198 patent. From the numerous examples given in the specification, it is clear that they did not make such a compound. This is confirmed by their statement that in the preferred compounds they made, x and y are each either 1 or 2. From the foregoing evidence of insurmountable difficulty I encountered in my attempts to produce such a compound by any known synthesis other than the ketoform synthesis, it is clear that those known synthesis could not have been used to make such a compound. Further, it is clear that the amine group in any of their compounds is an amine linkage which must always be a secondary amine --NH--; and the H in this linkage cannot be substituted.
Much later, in U.S. Pat. No. 4,452,884, Leppard disclosed combining a hydroxyphenyl group in which the 3,5-dialkyl substituents would produce a stable aroxyl radical, except that, having recognized that "A" in his structure could not be a disubstituted lone C atom (the alpha C atom), he specified that A is methylene or one of several groups having plural C atoms connecting the hydroxyphenyl group to the carbonyl C atom in his structure. In a subsequent patent (U.S. Pat. No. 4,518,679) Leppard et al disclosed many other piperidinyl derivatives linked to hydroxyphenyl groups with various linkages, none of which has the disubstituted alpha C atom connecting the phenyl ring to an amide carbonyl C atom. Thus it is evident that our 3,5-DHPIPA compounds are distinct and different from the hydroxyphenyl piperidyl compounds of both the '884 and the '679 Leppard patents.
It was in the foregoing framework of related aroxyl radicals with varying degrees of stability, measured as described hereinbelow, that I joined the search for a more stable aroxyl radical than the blue aroxyl radical, and sought to introduce the radical into a compound which might have enhanced stabilization activity.
Though I subscribed to the general notion that some bulking of the 1-C substituent would likely produce enhancement of the staibility of the bulked up aroxyl radical, relative to that of the blue aroxyl radical, there was no clear indication as to what might constitute the `proper` bulking. Many bulked up aroxyl radicals are less stable than the blue aroxyl, but there was no way of linking their activity to the lack of a disubstituted alpha-C atom. The disubstituted alpha-C atom was only present on the blue aroxyl radical in which the 3- and 5-carbons also had substituents which contained a disubstituted C-atom, and there was no particular reason to ascribe greater significance to the presence of such a C-atom in the substituent on the 1-C atom. Nor was there any reason to believe that an amide substituent might be about 50 times more effective than the blue aroxyl radical, and far more effective than other prior art `bulking` substituents, if it could be made to include a disubstituted alpha-C atom.
Apart from the inability to produce a disubstituted alpha C atom with a monohydroxyphenyl group in the structure, it is worth noting that in prior art hindered phenol hydroxyphenylalkleneyl isocyanurates which are antioxidants having superior effectiveness, for example, Irganox 1010, there is at least two C atoms connected to the 1-C atom of the hydroxyphenyl ring. It is this structure which in no small measure accounts for their effectiveness. As will be evident from data presented hereinafter, the compounds of our invention are inferior antioxidants, this property being attributable to the disubstituted alpha C atom; but they have a surprisingly superior stabilizing effect against degradation by ultraviolet light, attributable to the stability of the aroxyl radical formed.
Despite the activity of the 3,5-DHPIPA stabilizers of this invention, they would be of little value if they could not be made in good yields, that is, at least 50%. I found that a good yield of 3,5-DHPIPA was not obtained without one of the 3- or 5-alkyl substituents on the 3,5-DHPIPA being a tertiary alkyl substituent, which dictates one substituent on the 2,6-dialkyl phenol starting material. It is much preferred to have tert.-alkyl substituents on each of the 2- and 6-carbon atoms.